person
Author of the dev.to decision record, who describes himself as operating financial-sector Kubernetes clusters carrying 200 to 800 HPA objects.
Peering two Regions gets you a writable endpoint on each side and no failover step to run. The commit path pays an inter-Region round trip for it. The latency SLO that survived the migration is what breaks.
Reality
Auto-sync put whatever sat on disk into Git history, notebook outputs and stray credentials included. File-level staging fixes that, and it also breaks every handoff a team built on the assumption that saving was enough.
Reality
The 40x is arithmetic on an upstream default of five goroutines. It only transfers if every HPA evaluation really costs about 100ms. The reasoning that holds regardless is why a shared API server could not be handed the same setting.
Reality